CN111310788A - Water body pH value prediction method based on parameter optimization - Google Patents
Water body pH value prediction method based on parameter optimization Download PDFInfo
- Publication number
- CN111310788A CN111310788A CN202010041934.5A CN202010041934A CN111310788A CN 111310788 A CN111310788 A CN 111310788A CN 202010041934 A CN202010041934 A CN 202010041934A CN 111310788 A CN111310788 A CN 111310788A
- Authority
- CN
- China
- Prior art keywords
- value
- decision tree
- population
- candidate solution
- solution
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/24323—Tree-organised classifiers
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A20/00—Water conservation; Efficient water supply; Efficient water use
- Y02A20/152—Water filtration
Abstract
The invention discloses a water pH value prediction method based on parameter optimization. The improved whale optimization algorithm is applied to optimally design the training parameters of the gradient lifting decision tree, and then the optimally designed gradient lifting decision tree is utilized to realize the prediction of the pH value of the water body. In the improved whale optimization algorithm, a spiral motion operator, a surrounding motion operator and a combined jump operator are respectively executed to generate a new candidate solution, and the step size factor is adaptively adjusted, so that the searching performance of the algorithm is improved. The method can improve the prediction precision of the pH value of the water body.
Description
Technical Field
The invention relates to the field of water pollution treatment, in particular to a water pH value prediction method based on parameter optimization.
Background
With the continuous progress of human civilization, the living standard of people is gradually improved. However, in the development of the human society, the natural environment in which humans live is also subject to various pollutions. Among them, the problem of water pollution has become increasingly serious. Therefore, people pay more and more attention to the treatment of water pollution. In the process of treating water pollution, engineering technicians often need to master the dynamic change rule of the pH value of the water body, so as to plan a reasonable water pollution treatment scheme.
In recent years, machine learning techniques have been widely used in the human society. Machine learning techniques are continually being applied to the remediation of water pollution and are highly appreciated by engineering technicians. Gradient boosting decision trees are a very potential machine learning technique that has been successfully applied in many engineering practices. However, when the traditional gradient boost decision tree is applied to the prediction of the pH value of the water body, different training parameters are often set for specific engineering conditions to obtain a better prediction effect. And the setting of the training parameters of the traditional gradient boosting decision tree does not form systematic theoretical guidance. Therefore, when an engineer uses a traditional gradient lifting decision tree to solve the problem of predicting the pH value of the water body, the engineer often has the disadvantage of low prediction accuracy due to improper setting of training parameters.
Disclosure of Invention
The invention provides a water pH value prediction method based on parameter optimization. The improved whale optimization algorithm is utilized to carry out evolutionary optimization design on the training parameters of the gradient lifting decision tree, the defect that the traditional gradient lifting decision tree is low in prediction precision when used for predicting the pH value of the water body is overcome to a certain extent, and the prediction precision of the pH value of the water body can be improved.
The technical scheme of the invention is as follows: a water pH value prediction method based on parameter optimization comprises the following steps:
step 1, collecting a water quality sample data set;
step 2, preprocessing the collected water quality sample data set, and then dividing the collected water quality sample data set into a training data set and a testing data set;
step 3, determining an input variable and an output variable of the gradient lifting decision tree;
step 4, determining training parameters of optimal design required by the gradient lifting decision tree;
step 5, carrying out optimization design on training parameters of the gradient lifting decision tree by using an improved whale optimization algorithm;
step 6, forecasting the pH value of the water body by utilizing an optimally designed gradient lifting decision tree;
the optimized design of the training parameters of the gradient lifting decision tree by using the improved whale optimization algorithm in the step 5 comprises the following steps:
step 5.1, setting the population size WSize and the maximum iteration number MaxT;
step 5.2, setting the current iteration time t to be 0;
step 5.3, randomly generating WSize candidate solution composition population WP ═ { X1,X2,...,Xi,...,XWSizeIn which X isiIs the ith candidate solution in the population, and XiStoring training parameters of optimal design required by the gradient lifting decision tree; candidate solution subscript i ═ 1, 2., WSize;
step 5.4, setting the memory factor MKiRand (0,1), wherein rand (0,1) is represented at [0,1]]Randomly generating a real function;
step 5.5, calculating the adaptive value of each candidate solution in the population, finding out the candidate solution with the minimum adaptive value in the population and storing the candidate solution with the minimum adaptive value into the optimal solution XBest;
step 5.6, setting the motion flag mf to be 0, and then randomly generating a real number pec between [0,1 ];
step 5.7, if the pec is less than 0.5, go to step 5.10, otherwise go to step 5.8;
step 5.8, executing according to the spiral motion operator (1) to obtain a new candidate NU solutioni：
NUi＝|XBest-Xi|×exp(l)×cos(2×π×l)+XBest (1)
Wherein l is a random real number between [ -1,1 ]; pi is the circumference ratio; exp represents an exponential function with a natural constant e as the base; cos represents a cosine function;
step 5.9, go to step 5.17;
step 5.10, setting a motion mark mf to be 1; then calculating the disturbance amplitude SR according to the formula (2)i：
Wherein pr is a random real number between [0,1 ];
step 5.11, calculating the step factor AC according to the formula (3):
wherein ad is a convergence factor;
step 5.12, if the absolute value of the AC is less than 1, then go to step 5.13; otherwise go to step 5.15;
step 5.13, executing according to the surrounding motion operator (4) to obtain a new candidate solution NUi：
NUi＝XBest-AC×|2×fr×XBest-Xi| (4)
Wherein fr is a random real number between [0,1 ];
step 5.14, go to step 5.17;
step 5.15, calculating the skip factor LC according to the formula (5):
wherein, the log is a logarithmic function taking a natural constant e as a base number;
step 5.16, obtaining new candidate NU solution according to the execution of the combined jump operator (6)i：
Wherein WR is a short hop value; PR is the long jump value; kr is a random real number between [0,1 ]; DLow is the minimum value of candidate solutions in the population; DUp is the maximum value of candidate solutions in the population;
in the step 5.17, the step of the method,if new candidate solution NUiThe adaptive value of is better than that of XiUsing the new candidate solution NU in the populationiReplacement candidate solution XiOtherwise, the candidate solution X is keptiThe change is not changed;
step 5.18, if the motion identifier mf is equal to 1, go to step 5.19, otherwise go to step 5.20;
step 5.19, if the new candidate solves NUiThe adaptive value of is better than that of XiThe adaptation value of (A), then the memory factor MK is seti＝SRiOtherwise, the memory factor MK is keptiThe change is not changed;
step 5.20, finding out a candidate solution with the minimum adaptation value from the population, storing the candidate solution to the optimal solution XBest, and then setting the current iteration time t to be t + 1;
step 5.21, if the current iteration time t is greater than the maximum iteration time MaxT, go to step 5.22, otherwise go to step 5.6;
and 5.22, extracting the optimal design training parameters of the gradient lifting decision tree from the optimal solution XBest.
The improved whale optimization algorithm is applied to optimally design the training parameters of the gradient lifting decision tree, and then the optimally designed gradient lifting decision tree is utilized to realize the prediction of the pH value of the water body. In the improved whale optimization algorithm, an adaptive step factor adjusting strategy is designed, the step factor is dynamically adjusted according to feedback information of an adaptive value in a searching process, and the convergence speed of the whale optimization algorithm is accelerated; meanwhile, a combined jump operator combining short-distance jump and long-distance jump is designed to enhance the searching capability of the algorithm. The method can improve the prediction precision of the pH value of the water body.
Drawings
FIG. 1 is a flow chart of an improved whale optimization algorithm.
Detailed Description
The technical scheme of the invention is further specifically described by the following embodiments and the accompanying drawings.
Example (b):
in this embodiment, with reference to the accompanying drawings, the specific implementation steps of the present invention are as follows:
step 1, collecting a water quality sample data set, wherein the water quality sample data set comprises but is not limited to water quality data, and the water quality data comprises but is not limited to pH value, water temperature, dissolved oxygen, turbidity, conductivity and sulfide data;
step 2, preprocessing the collected water quality sample data set, and then dividing the collected water quality sample data set into a training data set and a testing data set; wherein preprocessing includes but is not limited to completion of missing data, deletion of abnormal data;
step 3, determining that the input variable of the gradient lifting decision tree is water quality data of one day: pH, water temperature, dissolved oxygen, turbidity, conductivity and sulfides, the output variables are: the pH value of the water body after two days; wherein the gradient boosting decision tree includes but is not limited to GradientBoosting regressor regression method in Python language machine learning algorithm tool software package scimit-lean;
step 4, determining training parameters of optimal design required by the gradient lifting decision tree as the maximum depth of the decision tree, the number of the decision trees and the learning rate; wherein the gradient boosting decision tree is obtained by integrating a plurality of decision trees;
step 5, carrying out optimization design on training parameters of the gradient lifting decision tree by using an improved whale optimization algorithm;
step 6, forecasting the pH value of the water body by utilizing an optimally designed gradient lifting decision tree;
the optimized design of the training parameters of the gradient lifting decision tree by using the improved whale optimization algorithm in the step 5 comprises the following steps:
step 5.1, setting the population size WSize to be 50 and the maximum iteration number MaxT to be 8000;
step 5.2, setting the current iteration time t to be 0;
step 5.3, randomly generating WSize candidate solution composition population WP ═ { X1,X2,...,Xi,...,XWSizeIn which X isiIs the ith candidate solution in the population, and XiThe training parameters of the optimal design required by the gradient lifting decision tree are stored: maximum depth of decision tree and decision treeThe number of and learning rate; candidate solution subscript i ═ 1, 2., WSize;
step 5.4, setting the memory factor MKiRand (0,1), wherein rand (0,1) is represented at [0,1]]Randomly generating a real function;
step 5.5, calculating the adaptive value of each candidate solution in the population, finding out the candidate solution with the minimum adaptive value in the population and storing the candidate solution with the minimum adaptive value into the optimal solution XBest; the calculation method of the adaptive value comprises the following steps: firstly, extracting the maximum depth of a decision tree, the number of the decision trees and the learning rate from a candidate solution, training a gradient lifting decision tree model on a training data set by using the obtained maximum depth of the decision tree, the number of the decision trees and the learning rate, then calculating the Mean Square Error (MSE) of the gradient lifting decision tree model on a test data set, and setting the obtained MSE as an adaptive value of the candidate solution; wherein, model represents a gradient lifting decision tree obtained by training;
step 5.6, setting the motion flag mf to be 0, and then randomly generating a real number pec between [0,1 ];
step 5.7, if the pec is less than 0.5, go to step 5.10, otherwise go to step 5.8;
step 5.8, executing according to the spiral motion operator (1) to obtain a new candidate NU solutioni：
NUi＝|XBest-Xi|×exp(l)×cos(2×π×l)+XBest (1)
Wherein l is a random real number between [ -1,1 ]; pi is the circumference ratio; exp represents an exponential function with a natural constant e as the base; cos represents a cosine function;
step 5.9, go to step 5.17;
step 5.10, setting a motion mark mf to be 1; then calculating the disturbance amplitude SR according to the formula (2)i：
Wherein pr is a random real number between [0,1 ];
step 5.11, calculating the step factor AC according to the formula (3):
wherein ad is a convergence factor;
step 5.12, if the absolute value of the AC is less than 1, then go to step 5.13; otherwise go to step 5.15;
step 5.13, executing according to the surrounding motion operator (4) to obtain a new candidate solution NUi：
NUi＝XBest-AC×|2×fr×XBest-Xi| (4)
Wherein fr is a random real number between [0,1 ];
step 5.14, go to step 5.17;
step 5.15, calculating the skip factor LC according to the formula (5):
wherein, the log is a logarithmic function taking a natural constant e as a base number;
step 5.16, obtaining new candidate NU solution according to the execution of the combined jump operator (6)i：
Wherein WR is a short hop value; PR is the long jump value; kr is a random real number between [0,1 ]; DLow is the minimum value of candidate solutions in the population; DUp is the maximum value of candidate solutions in the population;
step 5.17, if the new candidate is to solve NUiThe adaptive value of is better than that of XiUsing the new candidate solution NU in the populationiReplacement candidate solution XiOtherwise, the candidate solution X is keptiThe change is not changed;
step 5.18, if the motion identifier mf is equal to 1, go to step 5.19, otherwise go to step 5.20;
step 5.19, if the new candidate solves NUiThe adaptive value of is better than that of XiAn adaptation value ofSetting the memory factor MKi＝SRiOtherwise, the memory factor MK is keptiThe change is not changed;
step 5.20, finding out a candidate solution with the minimum adaptation value from the population, storing the candidate solution to the optimal solution XBest, and then setting the current iteration time t to be t + 1;
step 5.21, if the current iteration time t is greater than the maximum iteration time MaxT, go to step 5.22, otherwise go to step 5.6;
and 5.22, extracting the optimal design training parameters of the gradient lifting decision tree from the optimal solution XBest.
Claims (1)
1. The method for predicting the pH value of the water body based on parameter optimization is characterized by comprising the following steps of:
step 1, collecting a water quality sample data set;
step 2, preprocessing the collected water quality sample data set, and then dividing the collected water quality sample data set into a training data set and a testing data set;
step 3, determining an input variable and an output variable of the gradient lifting decision tree;
step 4, determining training parameters of optimal design required by the gradient lifting decision tree;
step 5, carrying out optimization design on training parameters of the gradient lifting decision tree by using an improved whale optimization algorithm;
step 6, forecasting the pH value of the water body by utilizing an optimally designed gradient lifting decision tree;
the optimized design of the training parameters of the gradient lifting decision tree by using the improved whale optimization algorithm in the step 5 comprises the following steps:
step 5.1, setting the population size WSize and the maximum iteration number MaxT;
step 5.2, setting the current iteration time t to be 0;
step 5.3, randomly generating WSize candidate solution composition population WP ═ { X1,X2,...,Xi,...,XWSizeIn which X isiIs the ith candidate solution in the population, and XiStoring the gradient boosting decision tree requirementsOptimizing the designed training parameters; candidate solution subscript i ═ 1, 2., WSize;
step 5.4, setting the memory factor MKiRand (0,1), wherein rand (0,1) is represented at [0,1]]Randomly generating a real function;
step 5.5, calculating the adaptive value of each candidate solution in the population, finding out the candidate solution with the minimum adaptive value in the population and storing the candidate solution with the minimum adaptive value into the optimal solution XBest;
step 5.6, setting the motion flag mf to be 0, and then randomly generating a real number pec between [0,1 ];
step 5.7, if the pec is less than 0.5, go to step 5.10, otherwise go to step 5.8;
step 5.8, executing the spiral motion operator (1) to obtain a new candidate NU solutioni：
NUi＝|XBest-Xi|×exp(l)×cos(2×π×l)+XBest (1)
Wherein l is a random real number between [ -1,1 ]; pi is the circumference ratio; exp represents an exponential function with a natural constant e as the base; cos represents a cosine function;
step 5.9, go to step 5.17;
step 5.10, setting a motion mark mf to be 1; then calculating the disturbance amplitude SR according to the formula (2)i：
Wherein pr is a random real number between [0,1 ];
step 5.11, calculating the step factor AC according to the formula (3):
wherein ad is a convergence factor;
step 5.12, if the absolute value of the AC is less than 1, then go to step 5.13; otherwise go to step 5.15;
step 5.13, executing the bounding motion operator (4) to obtain a new candidate solution NUi：
NUi＝XBest-AC×|2×fr×XBest-Xi| (4)
Wherein fr is a random real number between [0,1 ];
step 5.14, go to step 5.17;
step 5.15, calculating the skip factor LC according to the formula (5):
wherein, the log is a logarithmic function taking a natural constant e as a base number;
step 5.16, executing the combined jump operator (6) to obtain a new candidate solution NUi：
Wherein WR is a short hop value; PR is the long jump value; kr is a random real number between [0,1 ]; DLow is the minimum value of candidate solutions in the population; DUp is the maximum value of candidate solutions in the population;
step 5.17, if the new candidate is to solve NUiThe adaptive value of is better than that of XiUsing the new candidate solution NU in the populationiReplacement candidate solution XiOtherwise, the candidate solution X is keptiThe change is not changed;
step 5.18, if the motion identifier mf is equal to 1, go to step 5.19, otherwise go to step 5.20;
step 5.19, if the new candidate solves NUiThe adaptive value of is better than that of XiThe adaptation value of (A), then the memory factor MK is seti＝SRiOtherwise, the memory factor MK is keptiThe change is not changed;
step 5.20, finding out a candidate solution with the minimum adaptation value from the population, storing the candidate solution to the optimal solution XBest, and then setting the current iteration time t to be t + 1;
step 5.21, if the current iteration time t is greater than the maximum iteration time MaxT, go to step 5.22, otherwise go to step 5.6;
and 5.22, extracting the optimal design training parameters of the gradient lifting decision tree from the optimal solution XBest.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010041934.5A CN111310788B (en) | 2020-01-15 | 2020-01-15 | Water pH value prediction method based on parameter optimization |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010041934.5A CN111310788B (en) | 2020-01-15 | 2020-01-15 | Water pH value prediction method based on parameter optimization |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111310788A true CN111310788A (en) | 2020-06-19 |
| CN111310788B CN111310788B (en) | 2023-06-09 |
Family
ID=71161426
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010041934.5A Active CN111310788B (en) | 2020-01-15 | 2020-01-15 | Water pH value prediction method based on parameter optimization |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111310788B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113065687A (en) * | 2021-03-16 | 2021-07-02 | 西南民族大学 | Aquaculture method and system based on dissolved oxygen prediction |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2014157753A1 (en) * | 2013-03-28 | 2014-10-02 | 부산대학교 산학협력단 | System and method for providing water quality information capable of diagnosing and predicting state of water quality of water system |
| CN106124718A (en) * | 2016-06-27 | 2016-11-16 | 江西理工大学 | The Dissolved Oxygen in Water concentration prediction method that many strategy artificial bee colonies optimize |
| US20180158552A1 (en) * | 2016-12-01 | 2018-06-07 | University Of Southern California | Interpretable deep learning framework for mining and predictive modeling of health care data |
| CN109961186A (en) * | 2019-03-22 | 2019-07-02 | 大唐环境产业集团股份有限公司 | Desulphurization system operating parameter prediction technique based on decision tree and BP neural network |
| CN110110914A (en) * | 2019-04-26 | 2019-08-09 | 北京桃花岛信息技术有限公司 | Student's degree of difficulty prediction technique based on transformation decision tree and intelligent optimization method |
-
2020
- 2020-01-15 CN CN202010041934.5A patent/CN111310788B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2014157753A1 (en) * | 2013-03-28 | 2014-10-02 | 부산대학교 산학협력단 | System and method for providing water quality information capable of diagnosing and predicting state of water quality of water system |
| CN106124718A (en) * | 2016-06-27 | 2016-11-16 | 江西理工大学 | The Dissolved Oxygen in Water concentration prediction method that many strategy artificial bee colonies optimize |
| US20180158552A1 (en) * | 2016-12-01 | 2018-06-07 | University Of Southern California | Interpretable deep learning framework for mining and predictive modeling of health care data |
| CN109961186A (en) * | 2019-03-22 | 2019-07-02 | 大唐环境产业集团股份有限公司 | Desulphurization system operating parameter prediction technique based on decision tree and BP neural network |
| CN110110914A (en) * | 2019-04-26 | 2019-08-09 | 北京桃花岛信息技术有限公司 | Student's degree of difficulty prediction technique based on transformation decision tree and intelligent optimization method |
Non-Patent Citations (1)
| Title |
|---|
| 张宇航;项铁铭;王建成;: "基于维度变化的萤火虫优化算法" * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113065687A (en) * | 2021-03-16 | 2021-07-02 | 西南民族大学 | Aquaculture method and system based on dissolved oxygen prediction |
| CN113065687B (en) * | 2021-03-16 | 2023-09-05 | 西南民族大学 | Aquaculture method and system based on dissolved oxygen prediction |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111310788B (en) | 2023-06-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108803332B (en) | Path planning method based on improved biophysics | |
| CN103870647A (en) | Operation workshop scheduling modeling method based on genetic algorithm | |
| CN111832838B (en) | Method for predicting short-term wind power generation output power | |
| CN102968539A (en) | Method for massively and quickly generating format drawing | |
| CN112651576A (en) | Long-term wind power prediction method and device | |
| CN111310788B (en) | Water pH value prediction method based on parameter optimization | |
| CN116245019A (en) | Load prediction method, system, device and storage medium based on Bagging sampling and improved random forest algorithm | |
| CN112200316A (en) | GBDT learning method for online prediction task | |
| CN117093260B (en) | Fusion model website structure analysis method based on decision tree classification algorithm | |
| CN109214610B (en) | Saturated power load prediction method based on long-term and short-term memory neural network | |
| CN112907107B (en) | Fishery accident emergency treatment system and method based on multisource information fusion | |
| CN115879450B (en) | Gradual text generation method, system, computer equipment and storage medium | |
| CN117370568A (en) | Power grid main equipment knowledge graph completion method based on pre-training language model | |
| CN111931990A (en) | Power load prediction method based on optimized least square support vector machine | |
| CN110929809A (en) | Soft measurement method for key water quality index of sewage by using characteristic self-enhanced circulating neural network | |
| CN117786705B (en) | Statement-level vulnerability detection method and system based on heterogeneous graph transformation network | |
| Pencheva et al. | Modified multi-population genetic algorithms for parameter identification of yeast fed-batch cultivation | |
| Hashimoto et al. | Extraction of editing processes for modifying a program to another purpose based on representation learning | |
| CN117217150B (en) | DTCO formula modeling method based on genetic algorithm symbolic regression | |
| CN117875336A (en) | Household environment task decomposition method and system based on large-scale language model | |
| Eszes et al. | Applying Genetic Programming for the Inverse Lindenmayer Problem | |
| CN112257467B (en) | Decoding acceleration method of neural machine translation system for GPU (graphics processing Unit) equipment | |
| Wang et al. | ZIZO: A Novel Zoom-In–Zoom-Out Search Algorithm for the Global Parameters of Echo-State Networks | |
| CN114861609A (en) | Decision model generation method and device | |
| CN114184695A (en) | Parameter optimization-based method and system for predicting gas concentration in random forest oil |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |